fc8c76feb95fe5fb988441b6b42cb29be2369c0f,pynets/utils.py,,collect_pandas_df_make,#Any#Any#Any#Any#,217

Before Change

                    std_val = float(df_concatted_std[measure])
                    weighted_means.append((df_concat.loc[:, measures][measure] *
                                           np.array(1/(df_concat.loc[:, measures][measure])/std_val)).sum() /
                                          np.sum(np.array(1/(df_concat.loc[:, measures][measure])/std_val)))
                else:
                    weighted_means.append(np.nan)
            df_concatted_weight_means = pd.DataFrame(weighted_means).transpose()

After Change

            means = []
            for measure in measures:
                if measure in df_concatted_std.columns:
                    valsA = df_concat.loc[:, measures][measure].dropna()
                    valsB = valsA[(valsA.T != 0)]
                    means.append(np.nanmean(valsB))
                else:
                    means.append(np.nan)
            df_concatted_weight_means = pd.DataFrame(means).transpose()

Instances